Vassilis Nikolopoulos, Entrepreneur | Researcher | Innovator | Advisor: Analytics to engage
Consumer Engagement is a very hot space, either in the energy sectors where we do business, or in any other market-vertical (Retail, Banking, FinTech, Consumer goods, Support analytics, etc).
User Engagement is about engaging people, so you have to deal with a very difficult variable, which I would call "stochastic". People is a very tough variable to manage or to predict, so engagement is a multi-variable problem: we are talking about a mix of psychology, unique individuality, personalization, individual memories that lead to experience models, social interactions, influence metrics, sudden changes in feelings, a big time-variance and many more interconnected variables that can define or drive engagement. There are many models and theories deployed (ie. Social Norms, Personas, etc) however driving, measuring or even modeling/predicting human engagement and behaviors is still a big challenge that follows the rule: one size does not fill all; hence you need specific models and specific services to engage specific people and personas to specific vertical markets.
Engagement is engaging with something: a product, a digital journey, a participation or anything that has interaction and maybe produces value or/and revenues for the business cases (I take out social entrepreneurship). In order to achieve that, digital services and digital technology is being used and with the addition of Artificial Intelligence (AI) and advanced analytics (customer analytics), someone can analyze, generate insights, correlate and synthesize personalized applications and platforms to drive user engagement.
In order to engage or drive engagement, you need a complete methodology, a complete FrameWork (process, steps, actions, KPIs, next steps, trial&error techniques, etc), that can be deployed over a specific software; and because we are in the "Platformization" era, I would say you need a Platform.
We, at Intelen, have developed our own engagement framework for Utilities. This means that we have an extensive experience - ie R&D tech transfer methodologies, research background, metrics, user feedback, data, digital know how and other experienced-base skills acquired over a 4 year period of lean product development - capable to build and test digital engagement strategies and methods, enhanced with Game Mechanics. This framework can be used for other verticals as well (yes indeed, this is market proven and already adopted...), however I will focus right now on the utility vertical.
We had many data and insights from customers and deployments so far, revealing a need: that people want to engage with energy, utilities and energy-related layers (storage, EVs, energy sharing, etc) over a virtual "physical-digital" layer. In other words, a Physical-Digital Serious Game, interconnecting virtual digital gaming features with real sensor and IoT/smart meter/smart home/etc outputs. This is what we call DigiCal (Digital - Physical Interaction). This is a new incremental feature we will invest a lot in 2017, together with the AI (Artificial Intelligence) layer.
So far, based on our data and insights, we can see that our product (DiG) can raise NPS up to 20-25%, can reduce churn substantially, can make people interact with eachother (digital communities), can support and promote green mentality, can produce easily new revenue streams from customers, can engage users under a stable and consistent way and can produce loyal users that are having "fun engaging with us". This, according to BJ Fogg's Captology principle, can lead to a good recursive habit...
So, in order to quantify or predict engagement, you need to have first a model, a framework and tones of data. Because this is a Structural equation modeling (SEM) problem. As soon as you have the above, maybe with a statistical error you could be able to predict engagement, having as input variables: digital content, specific data sources, frequency of delivery, personas, game scenarios, loyalty indices and other interactions. This is a tough and difficult task indeed!
You can read the article here.